Fairness in representation: Quantifying stereotyping as a representational harm

Mohsen Abbasi; Sorelle A. Friedler; Carlos Scheidegger; Suresh Venkatasubramanian

doi:10.1137/1.9781611975673.90

Fairness in representation: Quantifying stereotyping as a representational harm

Mohsen Abbasi, Sorelle A. Friedler, Carlos Scheidegger, Suresh Venkatasubramanian

Computer Science

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

While harms of allocation have been increasingly studied as part of the subfield of algorithmic fairness, harms of representation have received considerably less attention. In this paper, we formalize two notions of stereotyping and show how they manifest in later allocative harms within the machine learning pipeline. We also propose mitigation strategies and demonstrate their effectiveness on synthetic datasets.

Original language	English (US)
Title of host publication	SIAM International Conference on Data Mining, SDM 2019
Publisher	Society for Industrial and Applied Mathematics Publications
Pages	801-809
Number of pages	9
ISBN (Electronic)	9781611975673
DOIs	https://doi.org/10.1137/1.9781611975673.90
State	Published - 2019
Event	19th SIAM International Conference on Data Mining, SDM 2019 - Calgary, Canada Duration: May 2 2019 → May 4 2019

Publication series

Name	SIAM International Conference on Data Mining, SDM 2019

Conference

Conference	19th SIAM International Conference on Data Mining, SDM 2019
Country	Canada
City	Calgary
Period	5/2/19 → 5/4/19

ASJC Scopus subject areas

Software

Access to Document

10.1137/1.9781611975673.90

Fingerprint Dive into the research topics of 'Fairness in representation: Quantifying stereotyping as a representational harm'. Together they form a unique fingerprint.

Cite this

Abbasi, M., Friedler, S. A., Scheidegger, C., & Venkatasubramanian, S. (2019). Fairness in representation: Quantifying stereotyping as a representational harm. In SIAM International Conference on Data Mining, SDM 2019 (pp. 801-809). (SIAM International Conference on Data Mining, SDM 2019). Society for Industrial and Applied Mathematics Publications. https://doi.org/10.1137/1.9781611975673.90

Fairness in representation : Quantifying stereotyping as a representational harm. / Abbasi, Mohsen; Friedler, Sorelle A.; Scheidegger, Carlos; Venkatasubramanian, Suresh.

SIAM International Conference on Data Mining, SDM 2019. Society for Industrial and Applied Mathematics Publications, 2019. p. 801-809 (SIAM International Conference on Data Mining, SDM 2019).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abbasi, M, Friedler, SA, Scheidegger, C & Venkatasubramanian, S 2019, Fairness in representation: Quantifying stereotyping as a representational harm. in SIAM International Conference on Data Mining, SDM 2019. SIAM International Conference on Data Mining, SDM 2019, Society for Industrial and Applied Mathematics Publications, pp. 801-809, 19th SIAM International Conference on Data Mining, SDM 2019, Calgary, Canada, 5/2/19. https://doi.org/10.1137/1.9781611975673.90

Abbasi M, Friedler SA, Scheidegger C, Venkatasubramanian S. Fairness in representation: Quantifying stereotyping as a representational harm. In SIAM International Conference on Data Mining, SDM 2019. Society for Industrial and Applied Mathematics Publications. 2019. p. 801-809. (SIAM International Conference on Data Mining, SDM 2019). https://doi.org/10.1137/1.9781611975673.90

@inproceedings{5acbc9bdd81749228a2a7d1afccfd339,

title = "Fairness in representation: Quantifying stereotyping as a representational harm",

abstract = "While harms of allocation have been increasingly studied as part of the subfield of algorithmic fairness, harms of representation have received considerably less attention. In this paper, we formalize two notions of stereotyping and show how they manifest in later allocative harms within the machine learning pipeline. We also propose mitigation strategies and demonstrate their effectiveness on synthetic datasets.",

author = "Mohsen Abbasi and Friedler, {Sorelle A.} and Carlos Scheidegger and Suresh Venkatasubramanian",

year = "2019",

doi = "10.1137/1.9781611975673.90",

language = "English (US)",

series = "SIAM International Conference on Data Mining, SDM 2019",

publisher = "Society for Industrial and Applied Mathematics Publications",

pages = "801--809",

booktitle = "SIAM International Conference on Data Mining, SDM 2019",

address = "United States",

note = "19th SIAM International Conference on Data Mining, SDM 2019 ; Conference date: 02-05-2019 Through 04-05-2019",

}

TY - GEN

T1 - Fairness in representation

T2 - 19th SIAM International Conference on Data Mining, SDM 2019

AU - Abbasi, Mohsen

AU - Friedler, Sorelle A.

AU - Scheidegger, Carlos

AU - Venkatasubramanian, Suresh

PY - 2019

Y1 - 2019

N2 - While harms of allocation have been increasingly studied as part of the subfield of algorithmic fairness, harms of representation have received considerably less attention. In this paper, we formalize two notions of stereotyping and show how they manifest in later allocative harms within the machine learning pipeline. We also propose mitigation strategies and demonstrate their effectiveness on synthetic datasets.

AB - While harms of allocation have been increasingly studied as part of the subfield of algorithmic fairness, harms of representation have received considerably less attention. In this paper, we formalize two notions of stereotyping and show how they manifest in later allocative harms within the machine learning pipeline. We also propose mitigation strategies and demonstrate their effectiveness on synthetic datasets.

UR - http://www.scopus.com/inward/record.url?scp=85066080818&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85066080818&partnerID=8YFLogxK

U2 - 10.1137/1.9781611975673.90

DO - 10.1137/1.9781611975673.90

M3 - Conference contribution

AN - SCOPUS:85066080818

T3 - SIAM International Conference on Data Mining, SDM 2019

SP - 801

EP - 809

BT - SIAM International Conference on Data Mining, SDM 2019

PB - Society for Industrial and Applied Mathematics Publications

Y2 - 2 May 2019 through 4 May 2019

ER -